ohi logo
OHI Science | Citation policy

1 Summary: Iconic Species Subgoal (Sense of Place)

This script prepares scores (status and trend) for Iconic Species in each global coastal region. For each iconic marine species, the countries of occurrence are pulled from the IUCN API. Extinction risk categories for each species are pulled based on current and past assessments; by tracking the assessed extinction risk over time, we can understand the trends of extinction risk for iconic species directly rather than using the “population trend” method from prior OHI assessments.

The Iconic Species sub-goal model calculates a region’s status based upon an unweighted average of species health for all ‘iconic’ species found within each reporting region.

From Halpern et al (2012):

Iconic species are those that are relevant to local cultural identity through a species’ relationship to one or more of the following: 1) traditional activities such as fishing, hunting or commerce; 2) local ethnic or religious practices; 3) existence value; and 4) locally-recognized aesthetic value (e.g., touristic attractions/common subjects for art such as whales). Habitat-forming species are not included in this definition of iconic species, nor are species that are harvested solely for economic or utilitarian purposes (even though they may be iconic to a sector or individual). …

Ultimately, almost any species can be iconic to someone, and so the intent with this goal was to focus on those species widely seen as iconic within a country, and iconic from a cultural or existence value (rather than for a livelihoods or extractive reason). …

The reference point is to have the risk status of all assessed species as Least Concern (i.e., a goal score = 1.0)

The Status of this sub-goal (XICO) is then the % of iconic species in each threat category (as defined by the IUCN Red List), such that:

\[X_{ICO} = \frac{\displaystyle\sum_{category}S_{cat}*w_{cat}}{\displaystyle\sum_{category}S_{cat}}\]

where for each IUCN threat category:

ICO trend is calculated in a similar manner, but weightings are assigned according to IUCN population trend: ‘Decreasing’ = -0.5, ‘Stable’ = 0.0, ‘Increasing’ = +0.5.

2 Updates from previous assessment

Changes since 2015 ICO subgoal for global OHI:

2.1 Future improvements?

  • Update list of iconic species… see issue #671

3 Data Sources

List of iconic species:

Species native country information:

4 Methods

Using the IUCN API, we accessed the full IUCN species list at http://apiv3.iucnredlist.org/api/v3/speciescount?token=. With some minor formatting, this list contains the following variables:

iucn_sid | kingdom | phylum | class | order | family | genus | sciname | population | category

4.1 get master list of Iconic Species

The list of Iconic Species is based upon the original ICO list generated in 2011, using species identified as globally iconic (WWF Flagship species and Priority species) or regionally iconic (based upon WWF regional/local priority species and nation-specific lists).

4.2 Identify countries with extant ICO species populations

Filtering the complete IUCN species list to include only the identified Iconic Species, we then use the IUCN API to access the list of countries in which each species occurs, from http://apiv3.iucnredlist.org/api/v3/species/countries/id/?token=. The country list identifies whether the species’ presence in that country is “Extant”, “Extinct Post-1500”, or “Possibly Extinct”; the “Extinct Post-1500” presence will be used later to identify locally extinct populations.

Sample of ico_spp_rgn saved at ~/github/ohiprep/globalprep/spp_ico/v2016/int/ico_spp_rgn.csv:

iucn_sid presence origin distribution_code sciname rgn_id rgn_name
2477 Extant Native Native Balaenoptera musculus 200 Angola
2477 Extant Native Native Balaenoptera musculus 172 Argentina
2477 Extant Native Native Balaenoptera musculus 16 Australia
2477 Extant Native Native Balaenoptera musculus 204 Bangladesh
2477 Extant Native Native Balaenoptera musculus 99 Benin
2477 Extant Native Native Balaenoptera musculus 108 Bermuda

4.3 Identify extinction risk from current and past assessments

We accessed the IUCN API to determine past IUCN assessments for each of the identified iconic species: http://apiv3.iucnredlist.org/api/v3/species/history/id/?token=.

Each assessment includes a year and an extinction risk, along with additional information on the assessment.

These raw assessments are cleaned up to standardize codes and categories, and category scores are assigned based on a scale from Least Concern = 1 to Extinct = 0, as per the Species OHI subgoal. Note that past assessments used different coding, especially prior to 1994; we reclassified older codes according to this chart:

New category <- original category/description                         
#     NT     <- "LOWER RISK/NEAR THREATENED (LR/NT)"                       
#      T     <- "THREATENED (T)" treat as "EN"
#     VU     <- "VULNERABLE (V)"                                           
#     EN     <- "ENDANGERED (E)"                                           
#  LR/CD     <- "LOWER RISK/CONSERVATION DEPENDENT (LR/CD)" treat as between VU and NT
#     CR     <- "VERY RARE AND BELIEVED TO BE DECREASING IN NUMBERS"       
#      T     <- "LESS RARE BUT BELIEVED TO BE THREATENED-REQUIRES WATCHING"
#     DD     <- "INSUFFICIENTLY KNOWN (K)"                                 
#     DD     <- "INDETERMINATE (I)"                                        
#     DD     <- "STATUS INADEQUATELY KNOWN-SURVEY REQUIRED OR DATA SOUGHT" 
#     NE     <- "NOT RECOGNIZED (NR)"

Using tidyr::complete() and tidyr::fill(), we create a full time series for all species from the earliest assessment to the most recent year.

4.4 Combine IUCN risk category time series with country <-> species lookup table

Using dplyr::full_join() we combine the ico_spp_rgn dataframe (iconic species by OHI region) with the ico_spp_cat dataframe (iconic species by category and year, with species info, year, and category info).

Sample of ico_spp_rgn_cat saved at ~/github/ohiprep/globalprep/spp_ico/v2016/int/ico_spp_rgn_cat.csv:

iucn_sid sciname year cat cat_score ico_gl ico_rgn_id presence origin distribution_code rgn_id rgn_name
2467 Balaena mysticetus 2000 CD 0.3 TRUE NA Extant Native Native 218 Canada
2467 Balaena mysticetus 2000 CD 0.3 TRUE NA Extant Native Native 145 Greenland
2467 Balaena mysticetus 2000 CD 0.3 TRUE NA Extant Native Native 163 United States
2467 Balaena mysticetus 2001 CD 0.3 TRUE NA Extant Native Native 218 Canada
2467 Balaena mysticetus 2001 CD 0.3 TRUE NA Extant Native Native 145 Greenland
2467 Balaena mysticetus 2001 CD 0.3 TRUE NA Extant Native Native 163 United States
2467 Balaena mysticetus 2002 CD 0.3 TRUE NA Extant Native Native 218 Canada
2467 Balaena mysticetus 2002 CD 0.3 TRUE NA Extant Native Native 145 Greenland
2467 Balaena mysticetus 2002 CD 0.3 TRUE NA Extant Native Native 163 United States
2467 Balaena mysticetus 2003 CD 0.3 TRUE NA Extant Native Native 218 Canada

4.5 Prep dataframe for toolbox; estimate status and trend

The toolbox wants rgn_id, species sciname, and extinction risk category for the basic calculations. Since some regions contain multiple subpops (or parent/subpop) we also include iucn_sid to differentiate. This information is included for each year, filtered back to the year 2000.

While the official calculations are performed in the toolbox, we perform the same basic calcs here to get a sense of the ICO status and trend ahead of time.

## Source: local data frame [0 x 5]
## 
## Variables not shown: rgn_id <int>, sciname <chr>, iucn_sid <int>, year
##   <int>, category <chr>.
## 
##    CD    CR    EN    EX    LC    NT    VU 
##   316  5776 10524    51  7113  8295 11717

4.5.1 Iconic Species full list (year == 2015)

4.5.2 Iconic Species processed status and trend by region (year == 2015)

4.6 Plot scores time series

To examine results of the new methods (including API-based data scraping and trend based on time series), we plot the estimated status and trend over time.

4.7 Compare scores OHI 2016 vs OHI 2015

To verify that this year’s methods are not too crazy relative to prior years’ methods, we plot the estimated status (for eez2013 scenario) based on v2016 methods and v2015 methods.

  • For the most part, the status scores are similar. Some differences can be attributed to:
    • updated region data from the new IUCN API
    • updated category data
    • improved methodology of including subpopulations
  • Trend scores are very different. This is clearly attributed to the time-series approach.
    • A linear model comparing v2015 trend to v2016 trend shows an adjusted R^2 = 0.3591, suggesting that the difference is more a matter of scaling factor and intercept.
    • While the differences are clear, the time-series approach is better as it more directly measures the change in category over time, as opposed to the simple “population trend” method which assigned +0.5, +0.0, -0.5 scores to increasing, stable, and decreasing populations per the most recent assessment.